The present invention relates generally to a dialogue complexity assessment method, and more particularly, but not by way of limitation, to a system, method, and computer program product for determining complexity as a data-driven, context-independent indicator to manage sets of dialogs and services operations.
Service industry thrives on engaged customers using a company's offerings, and dialogs, whether written or spoken, is a common form of such an interaction. Over time, organizations collect a sizable volume of dialogue data that may be proprietary or public depending on how customer service is provided.
As a customer calls up their service provider for a request, their interaction may be routine or extraordinary. Recently, there has been significant interest in the service management domain to automatically analyze such interaction data to better understand customer needs and ways to address them. For example, conventional techniques have considered tracking high-level indicators such as sentiments about how customer interactions are progressing in a service center and enable managers to take pro-active actions.
Thus, there is a need in the art for a dialogue complexity measure to characterize interactions with customers at the levels of utterances, turns and overall dialogs using dialogue data from online repositories as well as contact centers of service providers.